Pure Storage, Inc.

Brian Gold

John Martin Hayes

John D Davis

Shantanu R Gupta

Igor Ostrovsky

Zhangxi Tan

Robert H Lee

Ivan Jibaja

Roy Kim

Emily Watkins

Emily Potyraj

Hari Kannan

Ronald S Karr

Rusty Sears

John Colgrove

Ethan L Miller

Peter Vajgel

Joshua Robinson

Alex Ho

Joshua P Robinson

James Fisher

Bennett Amodio

Ramnath Sai Sagar Thumbavanam Padmanabhan

Taylor Ericson

John Hayes

John Davis

Generating a transformed dataset for use by a machine learning model in an artificial intelligence infrastructure that includes one or more storage systems and one or more graphical processing unit ('GPU') servers, including: storing, within one or more storage systems, a transformed dataset generated by applying one or more transformations to a dataset that are identified based on one or more expected input formats of data received as input data by one or more machine learning models to be executed on one or more servers; and transmitting, from the one or more storage systems to the one or more servers without reapplying the one or more transformations on the dataset, the transformed dataset including data in the one or more expected formats of data to be received as input data by the one or more machine learning models.

Optimizing dataset transformations for use by machine learning models

Improving machine learning models in an artificial intelligence infrastructure includes: storing, within one or more storage systems of an artificial intelligence infrastructure, information describing a dataset and one or more transformations applied to the dataset resulting in a transformed dataset; and storing, within the one or more storage systems, information describing only portions of previous versions of a machine learning model that differ from a current version of the machine learning model, wherein the previous versions used the transformed dataset as input during one or more prior executions by the artificial intelligence infrastructure.

Machine learning models in an artificial intelligence infrastructure

A method for a transactional commit in a storage unit is provided. The method includes receiving a logical record from a storage node into a transaction engine of a storage unit of the storage node and writing the logical record into a data structure of the transaction engine. The method includes writing, to a command queue of the transaction engine, an indication to perform an atomic update using the logical record and transferring each portion of the logical record from the data structure of the transaction engine to non-persistent memory of the storage unit as a committed transaction. A storage unit for a storage system is also provided.

Maintaining coherency in a distributed system

A method of applying a data format in a direct memory access transfer is provided. The method includes distributing user data throughout a plurality of storage nodes through erasure coding, wherein the plurality of storage nodes are housed within a single chassis that couples the storage nodes as a cluster, each of the plurality of storage nodes having nonvolatile solid-state memory for user data storage. The method includes reading a self-describing data portion from a first memory of the nonvolatile solid-state memory and extracting a destination from the self-describing data portion. The method includes writing data, from the self-describing data portion, to a second memory of the nonvolatile solid-state memory according to the destination.

Data storage in a zone drive

A method of applying an address space to data storage in a non-volatile solid-state storage is provided. The method includes receiving a plurality of portions of user data for storage in the non-volatile solid-state storage and assigning to each successive one of the plurality of portions of user data one of a plurality of sequential, nonrepeating addresses of an address space. The address range of the address space exceeds a maximum number of addresses expected to be applied during a lifespan of the non-volatile solid-state storage. The method includes writing each of the plurality of portions of user data to the non-volatile solid-state storage such that each of the plurality of portions of user data is identified and locatable for reading via the one of the plurality of sequential, nonrepeating addresses of the address space.

Distributed storage system

Enabling data portability between systems, including: submitting, to a cloud-based metadata store storing metadata describing data stored in a plurality of storage systems each coupled for data communications with a corresponding on-premises cloud infrastructure of a plurality of on-premises cloud infrastructures, a query; selecting a subset of the data responsive to the query and stored in at least one first storage system coupled to at least one first on-premises cloud infrastructure; and storing, from the at least one first storage system to a second storage system coupled to a second on-premises cloud infrastructure, the subset of the data responsive to the query.

Enabling data portability between systems

A method for updates in a storage system is provided. The method includes writing identifiers, associated with data to be stored, to storage units of the storage system and writing trim records indicative of identifiers that are allowed to not exist in the storage system to the storage units. The method includes determining whether stored data corresponding to records of identifiers is valid based on the records of the identifiers and the trim records.

Parallel update for a distributed system

A method for hardware assisted data lookup in a storage unit is provided. The method includes formatting data in at least one of a plurality of data formats for storage in the storage unit. The method includes configuring a logic unit with one or more parameters associated with the plurality of data formats and identifying incoming data with the one or more parameters as an instruction for execution.

Logic module for use with encoded instructions

A non-volatile solid-state storage is provided. The non-volatile solid state storage includes a non-volatile random access memory (NVRAM) addressable by a processor external to the non-volatile solid state storage. The NVRAM is configured to store user data and metadata relating to the user data. The non-volatile solid state storage includes a flash memory addressable by the processor. The flash memory is configured to store the user data responsive to the processor directing transfer of the user data from the NVRAM to the flash memory.

Erasure coded data within zoned drives

A method for storing data in a storage system having solid-state memory is provided. The method includes determining portions of the solid-state memory that have a faster access rate and portions of the solid-state memory that have a slower access rate, relative to each other or to a threshold. The method includes writing data bits of erasure coded data to the portions of the solid-state memory having the faster access rate, and writing one or more parity bits of the erasure coded data to the portions of the solid-state memory having the slower access rate. A storage system is also provided.

Optimize data protection layouts based on distributed flash wear leveling

Accelerating artificial intelligence workflows, including: receiving, from a computing process of an artificial intelligence workflow, a request for information stored on a data repository; issuing, from a user space of an operating system environment, parallel requests to the data repository using a network protocol that operates serially at the kernel level of the operating system environment; receiving, from the data repository, one or more responses to the parallel requests; and providing, to the computing process of the artificial intelligence workflow and based on the one or more responses to the parallel requests, a response to the request for information.

Accelerating artificial intelligence ('AI') workflows

Executing a machine learning model in an artificial intelligence infrastructure that includes one or more storage systems and one or more graphical processing unit ('GPU') servers, including: receiving, by a graphical processing unit ('GPU') server, a dataset transformed by a storage system that is external to the GPU server; and executing, by the GPU server, one or more machine learning algorithms using the transformed dataset as input.

Executing a machine learning model in an artificial intelligence infrastructure

Generating a transformed dataset for use by a machine learning model in an artificial intelligence infrastructure

Nonsequential readahead based on data access patterns, the method comprising: determining a set of access patterns for stored content; determining, based on the set of access patterns, a list of storage locations for content expected to be used; and prefetching, based on the list of storage locations for content expected to be used, one or more data objects.

Growing community of inventors

Mountain View, CA, United States of America

Brian Gold